Your First AI application¶

Going forward, AI algorithms will be incorporated into more and more everyday applications. For example, you might want to include an image classifier in a smart phone app. To do this, you'd use a deep learning model trained on hundreds of thousands of images as part of the overall application architecture. A large part of software development in the future will be using these types of models as common parts of applications.

In this project, you'll train an image classifier to recognize different species of flowers. You can imagine using something like this in a phone app that tells you the name of the flower your camera is looking at. In practice you'd train this classifier, then export it for use in your application. We'll be using this dataset from Oxford of 102 flower categories, you can see a few examples below.

The project is broken down into multiple steps:

  • Load the image dataset and create a pipeline.
  • Build and Train an image classifier on this dataset.
  • Use your trained model to perform inference on flower images.

We'll lead you through each part which you'll implement in Python.

When you've completed this project, you'll have an application that can be trained on any set of labeled images. Here your network will be learning about flowers and end up as a command line application. But, what you do with your new skills depends on your imagination and effort in building a dataset. For example, imagine an app where you take a picture of a car, it tells you what the make and model is, then looks up information about it. Go build your own dataset and make something new.

Install Datasets and Upgrade TensorFlow¶

To ensure we can download the latest version of the oxford_flowers102 dataset, let's first install both tensorflow-datasets and tfds-nightly.

  • tensorflow-datasets is the stable version that is released on a cadence of every few months
  • tfds-nightly is released every day and has the latest version of the datasets

We'll also upgrade TensorFlow to ensure we have a version that is compatible with the latest version of the dataset.

In [1]:
# %pip --no-cache-dir install tensorflow-datasets --user
# %pip --no-cache-dir install tfds-nightly --user
# %pip --no-cache-dir install --upgrade tensorflow --user

After the above installations have finished be sure to restart the kernel. You can do this by going to Kernel > Restart.

In [2]:
# %pip uninstall tensorflow -y
In [3]:
# %pip --no-cache-dir install --upgrade tensorflow-text 
In [4]:
# %pip install tf-nightly
In [5]:
%pip list | grep tensorflow
tensorflow                               2.12.0
tensorflow-addons                        0.21.0
tensorflow-cloud                         0.1.16
tensorflow-datasets                      4.9.2
tensorflow-decision-forests              1.4.0
tensorflow-estimator                     2.12.0
tensorflow-hub                           0.12.0
tensorflow-io                            0.32.0
tensorflow-io-gcs-filesystem             0.32.0
tensorflow-metadata                      0.14.0
tensorflow-probability                   0.20.1
tensorflow-serving-api                   2.12.1
tensorflow-text                          2.12.1
tensorflow-transform                     0.14.0
tensorflowjs                             3.15.0
Note: you may need to restart the kernel to use updated packages.
In [6]:
%pip show keras
Name: keras
Version: 2.12.0
Summary: Deep learning for humans.
Home-page: https://keras.io/
Author: Keras team
Author-email: keras-users@googlegroups.com
License: Apache 2.0
Location: /opt/conda/lib/python3.10/site-packages
Requires: 
Required-by: tensorflow
Note: you may need to restart the kernel to use updated packages.
In [7]:
# Import TensorFlow 
import tensorflow as tf
import tensorflow_datasets as tfds
import tensorflow_hub as hub

# Ignore some warnings that are not relevant (you can remove this if you prefer)
import warnings
warnings.filterwarnings('ignore')
/opt/conda/lib/python3.10/site-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.5
  warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}"
In [8]:
# TODO: Make all other necessary imports.
import json
import time

import matplotlib.pyplot as plt
import numpy as np
import logging
logger = tf.get_logger()
logger.setLevel(logging.ERROR)
In [9]:
print('Using:')
print('\t\u2022 TensorFlow version:', tf.__version__)
print('\t\u2022 tf.keras version:', tf.keras.__version__)
print('\t\u2022 Running on GPU' if tf.test.is_gpu_available() else '\t\u2022 GPU device not found. Running on CPU')
Using:
	• TensorFlow version: 2.12.0
	• tf.keras version: 2.12.0
	• Running on GPU
In [10]:
# Some other recommended settings:
%matplotlib inline
%config InlineBackend.figure_format = 'retina'
tfds.disable_progress_bar()

Load the Dataset¶

Here you'll use tensorflow_datasets to load the Oxford Flowers 102 dataset. This dataset has 3 splits: 'train', 'test', and 'validation'. You'll also need to make sure the training data is normalized and resized to 224x224 pixels as required by the pre-trained networks.

The validation and testing sets are used to measure the model's performance on data it hasn't seen yet, but you'll still need to normalize and resize the images to the appropriate size.

In [11]:
# import requests
# url = 'https://www.robots.ox.ac.uk/~vgg/data/flowers/102/102flowers.tgz'
# r = requests.get(url, allow_redirects=True)
# open('102flowers.tgz', 'wb').write(r.content)
In [12]:
# !tar zxvf  102flowers.tgz -C ./
In [13]:
# TODO: Load the dataset with TensorFlow Datasets. Hint: use tfds.load()
dataset, dataset_info = tfds.load('oxford_flowers102', as_supervised=True, with_info=True)
In [14]:
# Check that dataset is a dictionary
print('dataset has type:', type(dataset))

# Print the keys of the dataset dictionary
print('\nThe keys of dataset are:', list(dataset.keys()))
dataset has type: <class 'dict'>

The keys of dataset are: ['train', 'test', 'validation']
In [15]:
# TODO: Create a training set, a validation set and a test set.
training_set, validation_set, test_set = dataset['train'], dataset['test'], dataset['validation']

Explore the Dataset¶

In [16]:
# Display the dataset_info
dataset_info
Out[16]:
tfds.core.DatasetInfo(
    name='oxford_flowers102',
    full_name='oxford_flowers102/2.1.1',
    description="""
    The Oxford Flowers 102 dataset is a consistent of 102 flower categories commonly
    occurring in the United Kingdom. Each class consists of between 40 and 258
    images. The images have large scale, pose and light variations. In addition,
    there are categories that have large variations within the category and several
    very similar categories.
    
    The dataset is divided into a training set, a validation set and a test set. The
    training set and validation set each consist of 10 images per class (totalling
    1020 images each). The test set consists of the remaining 6149 images (minimum
    20 per class).
    
    Note: The dataset by default comes with a test size larger than the train size.
    For more info see this
    [issue](https://github.com/tensorflow/datasets/issues/3022).
    """,
    homepage='https://www.robots.ox.ac.uk/~vgg/data/flowers/102/',
    data_path='/root/tensorflow_datasets/oxford_flowers102/2.1.1',
    file_format=tfrecord,
    download_size=328.90 MiB,
    dataset_size=331.34 MiB,
    features=FeaturesDict({
        'file_name': Text(shape=(), dtype=string),
        'image': Image(shape=(None, None, 3), dtype=uint8),
        'label': ClassLabel(shape=(), dtype=int64, num_classes=102),
    }),
    supervised_keys=('image', 'label'),
    disable_shuffling=False,
    splits={
        'test': <SplitInfo num_examples=6149, num_shards=2>,
        'train': <SplitInfo num_examples=1020, num_shards=1>,
        'validation': <SplitInfo num_examples=1020, num_shards=1>,
    },
    citation="""@InProceedings{Nilsback08,
       author = "Nilsback, M-E. and Zisserman, A.",
       title = "Automated Flower Classification over a Large Number of Classes",
       booktitle = "Proceedings of the Indian Conference on Computer Vision, Graphics and Image Processing",
       year = "2008",
       month = "Dec"
    }""",
)
In [17]:
# TODO: Get the number of examples in each set from the dataset info.
num_training_examples  = dataset_info.splits['train'].num_examples
num_test_examples = dataset_info.splits['test'].num_examples
num_validation_examples = dataset_info.splits['validation'].num_examples
print('There are {:,} images in the training set'.format(num_training_examples))
print('There are {:,} images in the validation set'.format(num_validation_examples))
print('There are {:,} images in the test set'.format(num_test_examples))
There are 1,020 images in the training set
There are 1,020 images in the validation set
There are 6,149 images in the test set
In [18]:
# TODO: Get the number of classes in the dataset from the dataset info.
num_classes = dataset_info.features['label'].num_classes
num_classes
Out[18]:
102
In [19]:
# TODO: Print the shape and corresponding label of 3 images in the training set.
for image, label in training_set.take(3):
    print('The images in the training set have:\n\u2022 dtype:', image.dtype, '\n\u2022 shape:', image.shape)
The images in the training set have:
• dtype: <dtype: 'uint8'> 
• shape: (500, 667, 3)
The images in the training set have:
• dtype: <dtype: 'uint8'> 
• shape: (500, 666, 3)
The images in the training set have:
• dtype: <dtype: 'uint8'> 
• shape: (670, 500, 3)
In [20]:
# TODO: Plot 1 image from the training set. Set the title 
# of the plot to the corresponding image label. 
for image, label in training_set.take(1):
    image = image.numpy().squeeze()
    label = label.numpy()

plt.imshow(image, cmap= plt.cm.binary)
plt.colorbar()
plt.title(str(label))
plt.show()

Label Mapping¶

You'll also need to load in a mapping from label to category name. You can find this in the file label_map.json. It's a JSON object which you can read in with the json module. This will give you a dictionary mapping the integer coded labels to the actual names of the flowers.

In [21]:
with open('/kaggle/input/pytorch-challange-flower-dataset/cat_to_name.json', 'r') as f:
    class_names = json.load(f)
In [22]:
# TODO: Plot 1 image from the training set. Set the title 
# of the plot to the corresponding class name. 
for image, label in training_set.take(1):
    image = image.numpy().squeeze()
    label = label.numpy()

plt.imshow(image, cmap= plt.cm.binary)
plt.colorbar()
plt.title(class_names[str(label)])
plt.show()

Create Pipeline¶

In [23]:
# TODO: Create a pipeline for each set.
batch_size = 32
image_size = 224

def format_image(image, label):
    image = tf.cast(image, tf.float32)
    image = tf.image.resize(image, (image_size, image_size))
    image /= 255
    return image, label


training_batches = training_set.shuffle(num_training_examples//4).map(format_image).batch(batch_size).prefetch(1)
validation_batches = validation_set.map(format_image).batch(batch_size).prefetch(1)
testing_batches = test_set.map(format_image).batch(batch_size).prefetch(1)

Build and Train the Classifier¶

Now that the data is ready, it's time to build and train the classifier. You should use the MobileNet pre-trained model from TensorFlow Hub to get the image features. Build and train a new feed-forward classifier using those features.

We're going to leave this part up to you. If you want to talk through it with someone, chat with your fellow students!

Refer to the rubric for guidance on successfully completing this section. Things you'll need to do:

  • Load the MobileNet pre-trained network from TensorFlow Hub.
  • Define a new, untrained feed-forward network as a classifier.
  • Train the classifier.
  • Plot the loss and accuracy values achieved during training for the training and validation set.
  • Save your trained model as a Keras model.

We've left a cell open for you below, but use as many as you need. Our advice is to break the problem up into smaller parts you can run separately. Check that each part is doing what you expect, then move on to the next. You'll likely find that as you work through each part, you'll need to go back and modify your previous code. This is totally normal!

When training make sure you're updating only the weights of the feed-forward network. You should be able to get the validation accuracy above 70% if you build everything right.

Note for Workspace users: One important tip if you're using the workspace to run your code: To avoid having your workspace disconnect during the long-running tasks in this notebook, please read in the earlier page in this lesson called Intro to GPU Workspaces about Keeping Your Session Active. You'll want to include code from the workspace_utils.py module. Also, If your model is over 1 GB when saved as a checkpoint, there might be issues with saving backups in your workspace. If your saved checkpoint is larger than 1 GB (you can open a terminal and check with ls -lh), you should reduce the size of your hidden layers and train again.

In [24]:
# TODO: Build and train your network.
URL = "https://tfhub.dev/google/tf2-preview/mobilenet_v2/feature_vector/4"

feature_extractor = hub.KerasLayer(URL, input_shape=(image_size, image_size,3))
feature_extractor.trainable = False

model = tf.keras.Sequential([
        feature_extractor,
        tf.keras.layers.Dense(102, activation = 'softmax')
        ])

model.summary()
Model: "sequential"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 keras_layer (KerasLayer)    (None, 1280)              2257984   
                                                                 
 dense (Dense)               (None, 102)               130662    
                                                                 
=================================================================
Total params: 2,388,646
Trainable params: 130,662
Non-trainable params: 2,257,984
_________________________________________________________________
In [25]:
model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])
In [26]:
EPOCHS = 30
history = model.fit(training_batches,
                    epochs=EPOCHS,
                    validation_data=validation_batches)
Epoch 1/30
32/32 [==============================] - 31s 765ms/step - loss: 4.3341 - accuracy: 0.1078 - val_loss: 3.2340 - val_accuracy: 0.3601
Epoch 2/30
32/32 [==============================] - 23s 735ms/step - loss: 2.1133 - accuracy: 0.6814 - val_loss: 2.1239 - val_accuracy: 0.6160
Epoch 3/30
32/32 [==============================] - 19s 597ms/step - loss: 1.1318 - accuracy: 0.8863 - val_loss: 1.6648 - val_accuracy: 0.6936
Epoch 4/30
32/32 [==============================] - 18s 566ms/step - loss: 0.6797 - accuracy: 0.9627 - val_loss: 1.4087 - val_accuracy: 0.7352
Epoch 5/30
32/32 [==============================] - 19s 592ms/step - loss: 0.4481 - accuracy: 0.9873 - val_loss: 1.2756 - val_accuracy: 0.7430
Epoch 6/30
32/32 [==============================] - 19s 586ms/step - loss: 0.3154 - accuracy: 0.9941 - val_loss: 1.1768 - val_accuracy: 0.7601
Epoch 7/30
32/32 [==============================] - 23s 735ms/step - loss: 0.2381 - accuracy: 0.9961 - val_loss: 1.1202 - val_accuracy: 0.7603
Epoch 8/30
32/32 [==============================] - 23s 736ms/step - loss: 0.1844 - accuracy: 0.9980 - val_loss: 1.0780 - val_accuracy: 0.7674
Epoch 9/30
32/32 [==============================] - 18s 577ms/step - loss: 0.1487 - accuracy: 0.9990 - val_loss: 1.0374 - val_accuracy: 0.7697
Epoch 10/30
32/32 [==============================] - 23s 736ms/step - loss: 0.1208 - accuracy: 1.0000 - val_loss: 1.0064 - val_accuracy: 0.7738
Epoch 11/30
32/32 [==============================] - 18s 569ms/step - loss: 0.1006 - accuracy: 1.0000 - val_loss: 0.9856 - val_accuracy: 0.7757
Epoch 12/30
32/32 [==============================] - 24s 760ms/step - loss: 0.0856 - accuracy: 1.0000 - val_loss: 0.9687 - val_accuracy: 0.7803
Epoch 13/30
32/32 [==============================] - 19s 596ms/step - loss: 0.0742 - accuracy: 1.0000 - val_loss: 0.9539 - val_accuracy: 0.7783
Epoch 14/30
32/32 [==============================] - 23s 737ms/step - loss: 0.0647 - accuracy: 1.0000 - val_loss: 0.9409 - val_accuracy: 0.7774
Epoch 15/30
32/32 [==============================] - 18s 563ms/step - loss: 0.0571 - accuracy: 1.0000 - val_loss: 0.9311 - val_accuracy: 0.7783
Epoch 16/30
32/32 [==============================] - 19s 586ms/step - loss: 0.0510 - accuracy: 1.0000 - val_loss: 0.9181 - val_accuracy: 0.7816
Epoch 17/30
32/32 [==============================] - 18s 559ms/step - loss: 0.0459 - accuracy: 1.0000 - val_loss: 0.9117 - val_accuracy: 0.7816
Epoch 18/30
32/32 [==============================] - 23s 738ms/step - loss: 0.0414 - accuracy: 1.0000 - val_loss: 0.9023 - val_accuracy: 0.7827
Epoch 19/30
32/32 [==============================] - 19s 589ms/step - loss: 0.0378 - accuracy: 1.0000 - val_loss: 0.8975 - val_accuracy: 0.7826
Epoch 20/30
32/32 [==============================] - 18s 559ms/step - loss: 0.0345 - accuracy: 1.0000 - val_loss: 0.8909 - val_accuracy: 0.7834
Epoch 21/30
32/32 [==============================] - 19s 585ms/step - loss: 0.0318 - accuracy: 1.0000 - val_loss: 0.8861 - val_accuracy: 0.7842
Epoch 22/30
32/32 [==============================] - 18s 562ms/step - loss: 0.0294 - accuracy: 1.0000 - val_loss: 0.8763 - val_accuracy: 0.7857
Epoch 23/30
32/32 [==============================] - 23s 737ms/step - loss: 0.0272 - accuracy: 1.0000 - val_loss: 0.8754 - val_accuracy: 0.7848
Epoch 24/30
32/32 [==============================] - 18s 561ms/step - loss: 0.0253 - accuracy: 1.0000 - val_loss: 0.8699 - val_accuracy: 0.7848
Epoch 25/30
32/32 [==============================] - 19s 588ms/step - loss: 0.0236 - accuracy: 1.0000 - val_loss: 0.8683 - val_accuracy: 0.7857
Epoch 26/30
32/32 [==============================] - 18s 565ms/step - loss: 0.0221 - accuracy: 1.0000 - val_loss: 0.8635 - val_accuracy: 0.7853
Epoch 27/30
32/32 [==============================] - 19s 585ms/step - loss: 0.0207 - accuracy: 1.0000 - val_loss: 0.8590 - val_accuracy: 0.7866
Epoch 28/30
32/32 [==============================] - 23s 736ms/step - loss: 0.0195 - accuracy: 1.0000 - val_loss: 0.8565 - val_accuracy: 0.7870
Epoch 29/30
32/32 [==============================] - 23s 736ms/step - loss: 0.0183 - accuracy: 1.0000 - val_loss: 0.8554 - val_accuracy: 0.7871
Epoch 30/30
32/32 [==============================] - 23s 735ms/step - loss: 0.0173 - accuracy: 1.0000 - val_loss: 0.8513 - val_accuracy: 0.7889
In [27]:
# TODO: Plot the loss and accuracy values achieved during training for the training and validation set.
training_accuracy = history.history['accuracy']
validation_accuracy = history.history['val_accuracy']

training_loss = history.history['loss']
validation_loss = history.history['val_loss']

epochs_range=range(EPOCHS)

plt.figure(figsize=(8, 8))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, training_accuracy, label='Training Accuracy')
plt.plot(epochs_range, validation_accuracy, label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epochs_range, training_loss, label='Training Loss')
plt.plot(epochs_range, validation_loss, label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

Testing your Network¶

It's good practice to test your trained network on test data, images the network has never seen either in training or validation. This will give you a good estimate for the model's performance on completely new images. You should be able to reach around 70% accuracy on the test set if the model has been trained well.

In [28]:
# TODO: Print the loss and accuracy values achieved on the entire test set.
loss, accuracy = model.evaluate(testing_batches)

print('\nLoss on the TEST Set: {:,.3f}'.format(loss))
print('Accuracy on the TEST Set: {:.3%}'.format(accuracy))
32/32 [==============================] - 3s 79ms/step - loss: 0.7180 - accuracy: 0.8255

Loss on the TEST Set: 0.718
Accuracy on the TEST Set: 82.549%

Save the Model¶

Now that your network is trained, save the model so you can load it later for making inference. In the cell below save your model as a Keras model (i.e. save it as an HDF5 file).

In [29]:
# TODO: Save your trained model as a Keras model.
t = time.time()
saved_keras_model_filepath = './{}.h5'.format(int(t))
model.save(saved_keras_model_filepath)

Load the Keras Model¶

Load the Keras model you saved above.

In [30]:
# TODO: Load the Keras model
reloaded_keras_model = tf.keras.models.load_model(saved_keras_model_filepath, custom_objects={'KerasLayer':hub.KerasLayer})
reloaded_keras_model.summary()
Model: "sequential"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 keras_layer (KerasLayer)    (None, 1280)              2257984   
                                                                 
 dense (Dense)               (None, 102)               130662    
                                                                 
=================================================================
Total params: 2,388,646
Trainable params: 130,662
Non-trainable params: 2,257,984
_________________________________________________________________

Inference for Classification¶

Now you'll write a function that uses your trained network for inference. Write a function called predict that takes an image, a model, and then returns the top $K$ most likely class labels along with the probabilities. The function call should look like:

probs, classes = predict(image_path, model, top_k)

If top_k=5 the output of the predict function should be something like this:

probs, classes = predict(image_path, model, 5)
print(probs)
print(classes)
> [ 0.01558163  0.01541934  0.01452626  0.01443549  0.01407339]
> ['70', '3', '45', '62', '55']

Your predict function should use PIL to load the image from the given image_path. You can use the Image.open function to load the images. The Image.open() function returns an Image object. You can convert this Image object to a NumPy array by using the np.asarray() function.

The predict function will also need to handle pre-processing the input image such that it can be used by your model. We recommend you write a separate function called process_image that performs the pre-processing. You can then call the process_image function from the predict function.

Image Pre-processing¶

The process_image function should take in an image (in the form of a NumPy array) and return an image in the form of a NumPy array with shape (224, 224, 3).

First, you should convert your image into a TensorFlow Tensor and then resize it to the appropriate size using tf.image.resize.

Second, the pixel values of the input images are typically encoded as integers in the range 0-255, but the model expects the pixel values to be floats in the range 0-1. Therefore, you'll also need to normalize the pixel values.

Finally, convert your image back to a NumPy array using the .numpy() method.

In [31]:
# TODO: Create the process_image function
def process_image(image):
    image = tf.cast(image, tf.float32)
    image = tf.image.resize(image, (224, 224))
    image /= 255
    return image.numpy()

To check your process_image function we have provided 4 images in the ./test_images/ folder:

  • cautleya_spicata.jpg
  • hard-leaved_pocket_orchid.jpg
  • orange_dahlia.jpg
  • wild_pansy.jpg

The code below loads one of the above images using PIL and plots the original image alongside the image produced by your process_image function. If your process_image function works, the plotted image should be the correct size.

In [32]:
from PIL import Image

image_path = '/kaggle/input/test-images/test_images/hard-leaved_pocket_orchid.jpg'
im = Image.open(image_path)
test_image = np.asarray(im)

processed_test_image = process_image(test_image)

fig, (ax1, ax2) = plt.subplots(figsize=(10,10), ncols=2)
ax1.imshow(test_image)
ax1.set_title('Original Image')
ax2.imshow(processed_test_image)
ax2.set_title('Processed Image')
plt.tight_layout()
plt.show()

Once you can get images in the correct format, it's time to write the predict function for making inference with your model.

Inference¶

Remember, the predict function should take an image, a model, and then returns the top $K$ most likely class labels along with the probabilities. The function call should look like:

probs, classes = predict(image_path, model, top_k)

If top_k=5 the output of the predict function should be something like this:

probs, classes = predict(image_path, model, 5)
print(probs)
print(classes)
> [ 0.01558163  0.01541934  0.01452626  0.01443549  0.01407339]
> ['70', '3', '45', '62', '55']

Your predict function should use PIL to load the image from the given image_path. You can use the Image.open function to load the images. The Image.open() function returns an Image object. You can convert this Image object to a NumPy array by using the np.asarray() function.

Note: The image returned by the process_image function is a NumPy array with shape (224, 224, 3) but the model expects the input images to be of shape (1, 224, 224, 3). This extra dimension represents the batch size. We suggest you use the np.expand_dims() function to add the extra dimension.

In [33]:
# TODO: Create the predict function
def predict(image_path, model, top_k):
    image = Image.open(image_path)
    pre_image = np.asarray(image)
    processed_image = process_image(pre_image)
    processed_image = np.expand_dims(processed_image, axis = 0)
    predict_processed_image = model.predict(processed_image)
    values, indices = tf.math.top_k(predict_processed_image, k=top_k)
    probs = values.numpy()
    classes = indices.numpy() + 1
    return probs, classes

for image_batch, label_batch in testing_batches.take(1):
    prediction_1 = model.predict(image_batch)
    prediction_2 = reloaded_keras_model.predict(image_batch)
    difference = np.abs(prediction_1 - prediction_2)
    print(difference.max())
1/1 [==============================] - 0s 453ms/step
1/1 [==============================] - 0s 431ms/step
0.0

Sanity Check¶

It's always good to check the predictions made by your model to make sure they are correct. To check your predictions we have provided 4 images in the ./test_images/ folder:

  • cautleya_spicata.jpg
  • hard-leaved_pocket_orchid.jpg
  • orange_dahlia.jpg
  • wild_pansy.jpg

In the cell below use matplotlib to plot the input image alongside the probabilities for the top 5 classes predicted by your model. Plot the probabilities as a bar graph. The plot should look like this:

You can convert from the class integer labels to actual flower names using class_names.

In [34]:
# TODO: Plot the input image along with the top 5 classes
image_path = '/kaggle/input/test-images/test_images/hard-leaved_pocket_orchid.jpg'
image = Image.open(image_path)
image = np.asarray(image)
probs, classes = predict(image_path, model, top_k=5)
1/1 [==============================] - 1s 868ms/step
In [35]:
top_classes = [class_names[str(each)] for each in classes[0]]
In [36]:
top_classes
Out[36]:
['hard-leaved pocket orchid',
 'bearded iris',
 'anthurium',
 'giant white arum lily',
 'passion flower']
In [37]:
fig, (ax1, ax2) = plt.subplots(figsize=(10, 4), ncols=2)
ax1.imshow(image, cmap = plt.cm.binary)
ax1.axis('off')
ax1.set_title(top_classes[0])
ax2.barh(np.arange(5), probs[0])
ax2.set_aspect(0.1)
ax2.set_yticks(np.arange(5))
ax2.set_yticklabels(top_classes);
ax2.set_title('Class Probability')
ax2.set_xlim(0, 1.1)
plt.tight_layout()
In [ ]: